sign in sign up
<<
Home , Big gastrin, G cell, Secretin

J Clin Pathol: first published as 10.1136/jcp.s1-8.1.17 on 1 January 1978. Downloaded from

J. clin. Path., 33, Suppl. (Ass. Clin. Path.), 8, 17-25

Pathophysiology of and

KEITH D. BUCHANAN AND JOY E. S. ARDILL From the Department of Medicine, Queen's University of Belfast, Institute of Clinical Science, Grosvenor Road, Belfast BT12 6BJ

This paper will attempt to define the pathological THE NATURE OF GASTRIN roles of gastrin and secretin. It will deal predomin- Like many other gastrin exists in antly with gastrin which now has an accepted role in multiple molecular forms. The reason why peptide some diseases, whereas the pathological role of hormones exist in multiple forms is still uncertain secretin has still to emerge. Basic physiology and although in many instances, as in the case of pro- chemistry will only be described where these are and insulin, the relationship is likely to be considered to be relevant to the interpretation of the that ofa precursor and active . Preparations roles of these hormones in disease. of gastrin from gastrin-producing tumours, from normal tissues, and from circulating plasma have Gastrin been characterised. It is fortunate that in several conditions blood levels of gastrin are very high,

The history of gastrin began in 1905 when Edkins making investigation of circulating forms easier than copyright. observed than an extract of gastric antral mucosa in the case of some other peptide hormones. Al- proved to have extraordinary potency in stimulating though considerable progress in the characterisation secretion. Approximately 60 years later of tissue has been achieved it is still not gastrin was isolated and characterised by Gregory certain how closely these preparations resemble the and Tracy (1964). It is composed of a single peptide circulating species. chain of 17 amino-acids (G17) with a molecular Gregory and Tracy in 1975 isolated from both weight of 2100. Two forms of gastrin, I and II, were and hog antral mucosa two recognised initially, the only difference being that in containing 34 amino-acids. This gastrin has been http://jcp.bmj.com/ II the tyrosyl residue in position 12 is sulphated. referred to as G34 (big gastrin) in contrast with the The synthesis of human gastrin was achieved by heptadecapeptide G17 (small gastrin). Like G17 the Morley in 1968. two G34s differ only in the presence or absence of a The predominant biological activity of gastrin is sulphate on the tyrosyl residue. A 'mini-gastrin' has that of increasing gastric acid production. Several been isolated in small amounts from gastrinomas other roles have been assigned to it including trophic (Gregory and Tracy, 1975) and has been shown to effects on the , effects on the exocrine have the amino-acid composition of the COOH on September 29, 2021 by guest. Protected , and on the enteroinsular axis, but they do terminal tetradecapeptide (G14). Another variant not appear to play a role in pathological processes with the amino-acid composition of the 1-13 and will not be discussed further. fragment of G17 has been isolated from antral Gastrin is an antral hormone secreted by 'G cells' mucosa (Gregory, 1974). in the mid zone of the glands. However, smaller Components in the plasma which correspond amounts of gastrin are present in the and with these different gastrins have been identified in jejunum and, in some species, there may be small some instances (Rehfeld and Stadil, 1973). Com- amounts in the pancreas. Recently a substance which ponent m ofplasma corresponds to G17, component reacts with antigastrin antiserum has been detected II corresponds to G34, but component I of plasma, in the central nervous system (Vanderhaeghen et al., which elutes from Sephadex G50 between the void 1975). Immunochemical evidence so far suggests volume and G34, appears to have no tissue counter- that the 'brain gastrin' consists of COOH terminal part. An additional component in plasma, big big fragments of (Dockray, 1976). The gastrin (BBG), which elutes from Sephadex G50 in reason for the existence of gastrin-like peptides in the void volume, has been noted by Yalow and the central nervous system is not known and so far Berson (1972). It constitutes the major fraction of they have not been shown to play a role in any the immunoreactive gastrins extracted from fasting pathological process in man. plasma but does not increase with feeding; it seems 17 J Clin Pathol: first published as 10.1136/jcp.s1-8.1.17 on 1 January 1978. Downloaded from

18 Keith D. Buchanan and Joy E. S. Ardill likely to be an artefact which occurs on gel filtration by the groups of Gregory in Liverpool and Rehfeld of whole plasma as a result of non-specific binding in Copenhagen. In addition radioimmunoassay has ofgastrin peptides to plasma proteins, a phenomenon made it possible to measure many samples simul- observed frequently with other peptide hormones. taneously. However, there may be as yet unknown The amino-acid sequences of some of the gastrins species of gastrin which cannot be detected by are shown in Fig. 1 and the relationships between available antibodies, for example in large molecular tissue gastrins and plasma gastrins are illustrated in species of gastrin the areas recognised by the anti- Table 1. bodies may be hidden by other unknown and non- Plasma gastrin levels in fasting normal subjects do immunoreactive areas of the molecule. In addition not correlate with basal acid secretion rates. This is tiny fragments of gastrin could be present in the perhaps not surprising as, in the fasting state, BBG, circulation but unrecognised by any existing gastrin which is probably an artefact, constitutes the major antibody. part of the immunoreactive gastrin in plasma. This lack ofcorrelation may be partly because of different METHODS OF INVESTIGATION proportions of G34 and G17, as G17 has a shorter To assess the role of gastrin in health and disease half life than G34 in the circulation, though it is five the following techniques are required: radioimmuno- times more potent than G34 in stimulating acid assay, immunohistochemistry, electron microscopy secretion (Walsh et al., 1975). of G cells, and physical and chemical methods for Finally, it should be stated that the ability to the characterisation of gastrin, including gel detect multiple forms of a depends filtration and bioassay or assays using receptor site on the methods available. The considerable advances binding. Of these, radioimmunoassay has made the in the chemistry of gastrin have only been possible greatest contribution in recent years to the under- because of the use of multiple chemical techniques standing of gastrin pathophysiology. Gastrin can be assayed in plasma, serum, or tissue extracts. It iscopyright. relatively stable in blood, does not adhere to glass- Tissue gastrin Equivalent form ware, and can be assayed in plasma without pre- in plasma liminary extraction. The gastrin radioimmunoassay in use in our Big big gastrin Component I department was described by Ardill (1973). The Big gastrin (G34, non-sulphated and sulphated) Component II antibody 0098 was raised in rabbits against synthetic Gastrin (little gastrin) (G17) I non-sulphated Component III human gastrin I (Imperial Chemical Industries), http://jcp.bmj.com/ II sulphated containing amino-acids 2 to 17. The only other known Mini gastrin (a) G14 (residues 4 to 17 of G17) Component IV with which this assay (b) G13 (residues 1 to 13 of G17) cross reacts is cholecystokinin, 1:10 000 by weight. Table 1 Gastrins in human plasma and tissue The antibody also recognises gastrin G34 in equi- molar amounts. The gastrin is iodinated by the The plasma components still await full chemical identification, and only the probable tissue gastrin equivalent of the component is given. chloramine-T method and the product purified by

There is no tissue equivalent for 'big big gastrin' or component I. gel filtration. Equivalent amounts of plasma on September 29, 2021 by guest. Protected

1 2 3 4 57 6 7 19 1 11 12 13 14 1517 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 (a) Pyr - Leu-Gly -Pro-Gln-Gly- His- Pro- Ser- Leu-Vcl-AIa-Asp-Pro -Ser-Lys- Lys-Gln-Gly-Pro-Trp-Leu- Glu-Glu- Glu-Glu-Glu-Ala-Tyr-Gly -Trp-Met-Asp - Phe l l R NH2

1 2 3 4 5 6 7 8 9 10 1213 14 15 16 17 (b) Pyr-Gly-Pro-Trp -Leu -Glu-Glu- Gtu -Glu-Glu-Ata-Tyr-GIy-Trp -Met-Asp -Phe l I R NH2

1 2 3 4 5 6 7 8 9 10 11 12 B 14 (c) Trp- Leu-Glu-Glu-Glu-Glu-Ghu-Alo-Tyr-Gly1l -Trp-Met-Asp - Phe R NH2 Fig. 1 Amino-acid sequences ofhuman gastrins. (a) Human big gastrin (G34) MW 3839. (b) Human heptadecapeptide, little gastrin (G17, R = H or SOaH residues, 18-34 ofG34) MW 2098. (c) Huran minigastrin (G14, residues 21-34 of G34, 4-17 of G17). Pyr, pyroglutamyl; gastrin I, R = H; gastrin II, R = SOaH. J Clin Pathol: first published as 10.1136/jcp.s1-8.1.17 on 1 January 1978. Downloaded from

Pathophysiology ofgastrin and secrettin 19 adsorbed with charcoal (gastrin free) is added to blood gastrin estimation in the diagnosis ofduodenal the standards. The assay has a sensitivity of 5 ng/l ulcer (DU). Between June 1975 and February 1977, but this sensitivity can be improved by a variety of which was before the introduction of H2-receptor manipulations including the use of disequilibrium blocking drugs, gastrin assays were performed in conditions, reducing the concentrations of labelled 223 patients diagnosed provisionally as DU or hormone and antibody, and increasing the incubation Zollinger-Ellison syndrome (ZES). In 192 patients volume and the amount of sample assayed. Greater additional information was available and the final sensitivity would also be achieved if a more sophisti- diagnoses of these are shown in Table 2. Nine cated purification procedure were used for preparing subjects (4*7 %) had the ZES although some of them the labelled hormone, and also a less harsh iodination had had successful removal of the . A technique. However, the assay is highly satisfactory total of 147 patients had a duodenal ulcer but were for assay of gastrin in a variety of common clinical thought not to have a gastrinoma; 12 of these were conditions. considered to have another reason for hypergastrin- The results obtained by radioimmunoassay of aemia, for example hyperparathyroidism, post- gastrin will vary according to the specificity of the vagotomy, renal failure, and cimetidine therapy, antibody used. Antisera specific for different regions leaving 135 patients (70-3%) with duodenal ulcer of the molecule will react differently with the for further analysis. various molecular forms of gastrin. Some antisera, We regard the upper limit of normal of plasma like ours, will react with equimolar potency with gastrin as 96 ng/l, with a mean value of 36 ng/l; components I to m, whereas an antiserum reacting these values are derived from 132 normal subjects. exclusively with G17 (component m) has been All the patients who were eventually proved to have described by Dockray and Taylor (1976). More the ZES syndrome had levels between 540 and sophisticated analysis of the different gastrins in 6500 ng/l. All but one of the patients who had had

plasma can be achieved by combinations of gel successful removal of the tumour had gastrin levels copyright. filtration, tryptic cleavage, and radioimmunoassay, within the normal range, the exception having a as described by Rehfeld (1978). Clearly, there is the slightly raised value of 105 ng/l. possibility of raising antibodies which will be highly The plasma gastrin levels in the control group, specific for one or other species of gastrin, but before the DU group, and the patients with non-ulcer this can be achieved sufflcient amounts of a highly dyspepsia (NUD) are shown in Fig. 2. The patients purified preparation must be available. with DU had a mean fasting level of 52-6 ng/l

Although radioimmunoassay of gastrin is essential which is significantly raised compared with the http://jcp.bmj.com/ in the investigation of pathological states the value control group (P<0 0005). However, in patients of immunohistochemistry is less clear. The latter with NUD the mean (44-7 ng/l) was not significantly will make an important contribution to advances in different from the controls. Some of the blood knowledge of the role of gastrin in health and samples from the DU patients may not have been disease, but it generally contributes little to the taken under fasting conditions, but in 83 of them assessment of clinical problems, mainly because of known certainly to have been fasting, the mean the difficulty of quantitative estimation. However, was very group as a value similar to that ofthe whole. on September 29, 2021 by guest. Protected by means of it, the number of gastrin-secreting cells Of the 135 patients with duodenal ulcer unassoci- is shown to be grossly increased in achlorhydria, ated with other conditions, 20 had levels above the and it is also of considerable value in identifying gastrinomas, although a more sensitive and quantit- ative assessment is obtained by tissue extraction followed by radioimmunoassay of gastrin. Final diagnosis in 192 patients In clinical practice most states associated with Diagnosis No % abnormal gastrin secretion will be recognised by Duodenal ulcer (DU) 135 70.3 means of a robust and sensitive radioimmunoassay Pre- or postoperative ZES 9 4.7 DU with reason for hypergastrinaemia 12 6.2 which shows equimolar reactivity with the different Gastric ulcer 9 4.7 gastrin components in plasma and tissues. The use Gallstones 5 2.7 of the other to are Hiatus hernia 1 0.5 techniques referred above only Oesophageal reflux 1 0.5 required occasionally, but their overall contribution Disseminated sclerosis 1 0.5 to the future of research into gastrin in health and Non-ulcer dyspepsia 19 9.9 disease cannot be overemphasised. Table 2 Gastrin radioimmunoassay PLASMA GASTRIN ASSAY IN DUODENAL ULCER Final diagnosis in 192 patients whose blood samples were referred for Recently, we assessed the value of a single fasting gastrin assay with the provisional diagnosis of '?DU' or '?ZES'. J Clin Pathol: first published as 10.1136/jcp.s1-8.1.17 on 1 January 1978. Downloaded from

20 Keith D. Buchananz and Joy E. S. Ardill 60- Hypergastrinaemic D U subjects _ 9 280- c 40-

a - 240- E 20- n = 132 n =135 n= 19 n= 83 U) 200- Z 5 0 . c Cons DU NUD DU fast 0- 160- P< 0 0005 NS P< 0.0025 c Fig. 2 Plasma gastrin levels (mean + SEM) in control Un subjects (Cons), all patients with duodenial ulcer (DU), a 0 non-uilcer dyspepsia (NUD), and knownt fastiing patients W 120- sw.ith duodenal iulcer (DUfast). E U) 0 normal range. The results of another analysis per- i~ 80 - formed in 13 of these are shown in Fig. 3. They were almost all lower than the first estimation, but several remained above the normal range. The 20 40-

DU subjects with hypergastrinaemia were carefully copyright. analysed in order to discover whether they could form a separate group from the duodenal ulcer O- v 4 4 subjects with normal plasma gastrin levels. Basal Cons Sample 1 Repeat acid output and maximal acid output after penta- Fig. 3 Difference between gastrin levels in 2 samples gastrin were 3-09 and 23-81 mmol/h, respectively, ofbloodfrom 20 patients with DUfound to have which were less than the corresponding values of high levels in the first sample in the absence ofknown

5 70 and 30-78 for the DU group as a whole. Three causes ofhypergastrinaemia (such as gastrinoma, http://jcp.bmj.com/ subjects, however, did have pyloric stenosis and renalfailure, etc). Thre range for control subjects there were three children aged 8, 10, and 13 years, (Cons) is also shown. one of whom had a jejunal ulcer. In some of these subjects other tests were performed including antral duodenal ulcer is only of value in identifying the biopsy, with radioimmunoassay of gastrin in an patients who have the Zollinger-Ellison syndrome. antral extract, immunohistochemical assessment of There is no evidence that gastrin plays an im- G cells, and provocative tests for gastrin release. portant role in the pathogenesis of gastric ulcer. on September 29, 2021 by guest. Protected However, no evidence was adduced that any of these Many patients with gastric ulcer have hypochlor- subjects differed from the other DU subjects. hydria which could account for a modest rise of Our data showing that basal plasma gastrin is plasma gastrin in some patients (Trudeau and raised in patients with duodenal ulcer differ from McGuigan, 1971). However, some patients with those of most other authors, who report normal gastric ulcer have antral gastritis and this might levels (Walsh and Grossman, 1973), although higher impair the release of gastrin. levels in the DU group than in a control population have been reported after a protein meal (McGuigan HYPERGASTRINAEMIA and Trudeau, 1973). Raised levels in patients with The usual causes of hypergastrinaemia are shown in duodenal ulcer were reported by Byrnes etal. (1970), Table 3, and most of them will be easily identified. but their assay and normal range differ greatly from The commonest reason for hypergastrinaemia in our ours. experience is achlorhydria, hypergastrinaemia being Although the mean plasma gastrin level in the presumably the result of the lack of the normal duodenal ulcer group is significantly higher than in feedback inhibition of acid on gastrin release (Fig. the controls, the assay is nevertheless not useful 4). There is usually a massive rise of plasma gastrin diagnostically, as in only 20 of the DU group was in achlorhydria, the highest levels occurring in the level above the normal range. It is concluded pernicious anaemia. The cause of the hypergastrin- that plasma gastrin estimation in patients with aemia can easily be confirmed by estimating the J Clin Pathol: first published as 10.1136/jcp.s1-8.1.17 on 1 January 1978. Downloaded from

Pathophysiology ofgastrin and secretin 21

Gastrinoma is short, and often there are numerous complications Achlorhydria including pyloric stenosis and perforation; fre- Renal failure quently several operations have been performed Antral hyperplasia Vagotomy before the correct diagnosis has been made. How- Retained isolated antrum ever, the clinical picture is not so distinctive as to Short gut syndrome Cimetidine therapy make the diagnosis easy. Indeed the diagnosis is usually made by laparotomy, but if the correct Table 3 Causes ofhypergastrinaemia diagnostic approach is adopted the surgeon will not find himself unexpectedly confronted with a tumour at laparotomy. 1600 In addition to radiological investigations, the two laboratory procedures which are crucial in the diagnosis of gastrinoma are the assessment of basal gastric acid secretion (normally up to 15 mmol/h) and the fasting plasma gastrin level. We have always found plasma gastrin levels to be clearly above the normal range in patients with gastrinoma and regard further stimulatory tests of gastrin release, such as a protein meal, intravenous secretin, , or calcium (Basso and Passaro, 1970; Hansky et al., 1971; Creutzfeldt et al., 1975), as generally un- necessary. However, they may occasionally be useful when the rise of plasma gastrin is not as great as expected in gastrinoma. It is said that in patients copyright. with gastrinomas gastrin levels rise after the intra- venous stimuli, whereas in normal subjects they do not. The question arises of whether plasma gastrin should be measured in every patient who has peptic Controls P A Achlorhydria ulceration. The radioimmunoassay of plasma

n=100 n= 50 n =50 gastrin is easy to perform and we believe that one http://jcp.bmj.com/ Mean ± SE technician could easily cope with a single blood sample from every patient with peptic ulceration in Fig. 4 Fasting plasma gastrin levels in condtros and a population of 2 million, and that this would result in patients with pernicious anaemia (PA) antd in early diagnosis of gastrinoma and a considerable achlorhydria without anaemia. decline in the morbidity and mortality. A consider- able proportion of patients with gastrinoma have

gastric acidity or by showing circulatirng antibodies multiple endocrine adenomatosis (4 of 11 seen in on September 29, 2021 by guest. Protected to the parietal cells, and there is usuall) no difficulty this department in 8 years) and it is therefore in distinguishing gastrinoma from achllorhydria. important to screen these patients for other endocrine Basal plasma gastrin levels and those after tumours. The estimation of plasma pancreatic stimulation by food are increased by H2-receptor peptide (page 43) may also be a useful adjunct in antagonist therapy (Buchanan et al., 1978). The diagnosis. The treatment of gastrinoma is discussed importance of this hypergastrinaemia in long term elsewhere by Professor Welbourn (page 85). cimetidine therapy is unknown but it raises the possibility of an increase in the pariettal cell mass. ANTRAL G CELL HYPERPLASIA It is clearly important when assessing a high plasma Cowley et al. (1973) and Ganguli et al. (1974) gastrin level and acid secreting status in a patient described the syndrome of hyperplasia of the antral with duodenal ulcer to know whether the patient is gastrin-secreting cells in . It was taking cimetidine, which may have to t e withdrawn suggested that there was a separate entity of patients for an accurate assessment to be made. with peptic ulcer and hypersecretion of acid who did not have a gastrinoma, but who had high fasting GASTRINOMA (ZOLLINGER-ELLISON SYNDROME) plasma gastrin levels and a further increase after a Patients with a gastrinoma usually have severe protein meal. Such patients had increased numbers progressive ulcer disease with multii ple ulcers in of G cells in the antrum. A new classification of the atypical sites (Isenberg et al., 1973). The history ZES was developed, patients with antral G cell J Clin Pathol: first published as 10.1136/jcp.s1-8.1.17 on 1 January 1978. Downloaded from

22 Keith D. Buchanan and Joy E. S. Ardill hyperplasia being referred to as ZES type I, and as it has not yet found an important place in clinical those with the classical ZES (gastrinoma) were medicine. called ZES type HI. However, this has not been Secretin was the first hormone to be described. universally accepted though it is easier to confirm In 1902, Bayliss and Starling found that when an the existence of a new syndrome than to refute it. It extract of the upper intestinal mucosa was injected is evident, however, that the syndrome must be very intravenously into anaesthetised dogs, the flow of rare. We have not seen such a case and Hansky was stimulated. They concluded (1974) found only three to conform to Ganguli's that an active substance was secreted from the criteria among more than 400 patients with recurrent duodenum into the blood stream to act subsequently ulcer after surgery or peptic ulcer with hyper- on the exocrine pancreas, and they called it 'secretin'. gastrinaemia. There are several reasons which may In 1966, Mutt and Jorpes described the amino- account for this difficulty. The antral G cells have a acid sequence of secretin, which is composed of 27 patchy distribution, are often affected by local amino-acid residues and has structural similarities pathology such as gastritis (McFarland et al., 1978b), to glucagon, vasoactive intestinal peptide, and and quantitative assessment of their number is gastric inhibitory polypeptide.' It was synthetised by difficult. In addition, there are wide variations Bodansky et al. (1966). The hormone is present in between people in the gastrin content of the antrum both the duodenum and jejunum. (McFarland et al., 1978b). We have been unable to The development of a radioimmunoassay for find any clear cut differences in antral gastrin between secretin was somewhat delayed because of the lack patients with duodenal ulcer and non-ulcer dyspep- of pure secretin. Antibodies to secretin are easily sia, although a considerable increase in antral produced, but labelling of the hormone was difficult gastrin concentration has been shown in patients with until the advent of synthetic secretin (Holohan et al., pernicious anaemia. 1973). The entity of G cell hyperplasia is appealing. If Secretin radioimmunoassays are highly specificcopyright. this condition were easy to diagnose it could lead to and do not cross react with other known peptides. rational therapy for such patients, that is antrectomy. Most authors, using gel filtration and radioimmuno- However, it is probably very rare. assays, have reported a single species of secretin in tissues but we have found a degree of heterogeneity GASTRIN DEFICIENCY in both tissues (Mason et al., 1977) and plasma Pathological states associated with abnormalities of (Mason et al., 1979). The reason for this discrepancy gut hormones are usually ascribed to excessive is not apparent though there have been some http://jcp.bmj.com/ production of the hormone either from hyperplastic differences in the technique of immunopurification cells or from tumours. Seldom is a clinical condition procedures. Differences in the nature of the anti- ascribed to deficient production of the hormone, bodies are a possible explanation, but seem unlikely, although this does not preclude the occurrence of as nearly all antibodies to secretin in our laboratory such a deficiency. It is difficult to produce a clinical are directed towards the C terminal region of state of deficiency surgically because of the diffuse secretin, so that it is likely that the antisera used by and widespread nature of this . other authors are also C terminal (Boden and Chey, on September 29, 2021 by guest. Protected Recently, however, a patient with pernicious 1973). Nevertheless, antibodies directed against anaemia was found to have plasma gastrin levels other parts of the secretin molecule could account within the normal range (McFarland et al., 1978a) for differences between laboratories. instead of the expected high levels. The patient had It appears likely that secretin has a physiological autoimmune disease with diabetes mellitus, and role in stimulating the exocrine pancreas to secrete gastric and antibodies. The a juice with a low content and a high gastrin response to a protein meal was subnormal content. The exocrine pancreas is and antral biopsy showed a normal histological exquisitely sensitive to small physiological doses of appearance, but gastrin could not be shown in the secretin given intravenously (Hacki et al., 1977). biopsy material either by radioimmunoassay or by Most would agree that secretin is released after immunohistochemistry. We have been unable to instillation of acid into the duodenum (Boden et a!., determine whether this patient has autoantibodies 1974), but this may not represent a physiological directed towards the gastrin cells. stimulus. Chey et al. (1977) have reported that plasma secretin rises in the postprandial state in Secretm man, but others have been unable to show this (Bodn, 1978). We ourselves, using a sensitive Only a brief summary of the present status of radioimmunoassay with a detection limit of 3 ng/l, secretin in physiology and pathology will be given lGlucose-dependent insulin-releasing polypeptide. J Clin Pathol: first published as 10.1136/jcp.s1-8.1.17 on 1 January 1978. Downloaded from

Pathophysiology ofgastrin and secretin 23 were unable to show rises in plasma secretin after Zollinger-Ellison syndrome resulting from a gastri- the ingestion of a variety of oral stimuli including noma. Radioimmunoassay of gastrin in plasma has , fat, protein, alcohol, , and been the most important advance in technology and a standard meal. However, Schaffalitzky de has allowed a clearer definition of the role of gastrin Muckadell and Fahrenkrug (1978) have shown that in health and disease. Our own studies provide no brief rises in plasma secretin concentration coincide evidence for the existence of a hypergastrinaemic with rapid falls in intraduodenal pH in the fasting subgroup of DU patients (without gastrinoma) and state as well as postprandially. It seems possible, we recommend caution in the diagnosis of antral G therefore, that such rapid fluctuations may account cell hyperplasia. Plasma gastrin radioimmunoassay for the failure to show a rise in plasma secretin when is an easy screening procedure for gastrinoma and measurements are made at arbitrary intervals. its widespread application could result in consider- However, it is clear that plasma secretin does not able improvement in morbidity and mortality in rise after a meal to the same extent as other gut this condition. peptides, such as gastrin, gastric inhibitory poly- Although secretin may have been the first hormone peptide, insulin, pancreatic peptide, etc. to be described, its roles in physiology and pathology Because of the lack of rise of secretin after a meal in man still await definition. Although it is released and because secretin has lipolytic activity, we looked by intraduodenal acid, and although tiny physio- for a possible metabolic role. We found marked logical amounts given intravenously will stimulate rises of plasma secretin during starvation in normal the exocrine pancreas, it is only with great difficulty men (Henry et al., 1975), though others have been that rises of plasma secretin are detected after a meal. unable to confirm this (Greenberg and Bloom, 1978). A considerable amount of controversy surrounds Subsequent studies in our laboratory have confirmed certain issues with respect to secretin and it is clearly the original observation and shown that the secretin required to define the specificity of secretin radio- found in the plasma during starvation was identical immunoassays and also to establish the relation- copyright. with that found after a single overnight fast (Mason ship of secretin in plasma to that in tissues. et al., 1979). Clearly there is insufficient evidence at present to define the role of secretin in physiology. There are References some unsolved problems concerning the specificity of the radioimmunoassay, which require for their Ardill, J. E. S. (1973). The Measurement of Gastrin by solution a similarly exhaustive approach as applied Radioimmunoassay. Ph.D. Thesis, Queen's University http://jcp.bmj.com/ to the study of gastrin (see contribution by Professor of Belfast. Rehfeld, page 26). Basso, N., and Passaro, F., Jr. (1970). Calcium- stimulated gastric secretion in the Zollinger-Ellison syndrome. Archives ofSurgery, 101, 399-402. THE ROLE OF SECRETIN IN PATHOLOGY Bayliss, W. M., and Starling, E. H. (1902). The mechanism The secretion of secretin by a tumour has never been of pancreatic secretion. Journal of Physiology, 28, described. Therefore, the clinical and pathological 325-353. changes that might be associated with excessive Bloom, S. R., Patel, H. R., and Johnston, D. I. (1976). on September 29, 2021 by guest. Protected secretion of secretin are uncertain. Raised levels of Failure of secretin release in coeliac disease (Abst.). circulating secretin have been reported in renal Gut, 17, 812. failure (Rhodes et al., 1975b), but have not been Bloom, S. R., and Ward, A. S. (1975). Failure of secretin confirmed by us (unpublished observations). release in patients with duodenal ulcer. British Medical Impaired release of secretin in response to the Journal, 1, 126-127. ingestion or intraduodenal infusion of acid has been Bodansky, M., Ondetti, M. A., Levine, S. D., Narayanan, V. L., von Saltza, M., Sheehan, J. T., Williams, N. J., reported in coeliac disease (Rhodes et al., 1975a; and Sabo, E. F. (1966). Synthesis of a heptacosapep- Bloom et al., 1976; O'Connor et al., 1977), and in tide amide with the hormonal activity of secretin. patients with duodenal ulcer (Bloom and Ward, Chemistry and Industry, 1757-1758. 1975), although the latter was not confirmed in more Boden, G. (1978). The secretin assay. In Gut Hormones, recent studies (Isenberg et al., 1977; McLoughlin et pp. 169-175, ed S. R. Bloom. Churchill Livingstone, al., 1978). Edinburgh. Boden, G., and Chey, W. Y. (1973). Preparation and Conclusion specificity of antiserum to synthetic secretin and its use in a radioimmunoassay (RIA). , 93, 1617-1624. Gastrin has a clear physiological role in acid secretion Boden, G., Essa, N., Owen, 0. E., and Reichle, F. A. in man. It is implicated in several pathological (1974). Effects of intraduodenal administration of states, but the most important of these is the HCI and glucose on circulating immunoreactive J Clin Pathol: first published as 10.1136/jcp.s1-8.1.17 on 1 January 1978. Downloaded from

24 Keith D. Buchanan and Joy E. S. Ardill secretin and insulin concentrations. Journal of Clinical duodenum on pancreatic bicarbonate secretion and Investigation, 53, 1185-1193. plasma secretin in duodenal ulcer patients and normal Buchanan, K. D., Spencer, A., Ardill, J., and Kennedy, subjects. Gastroenterology, 72, 6-8. T. L. (1978). Hypergastrinaemia due to cimetidine Isenberg, J. I., Walsh, J. H., and Grossman, M. 1. (1973). (Abst.). Gut, 19, A437. Zollinger-Ellison syndrome. Gastroenterology, 65, Byrnes, D. J., Young, J. D., Chisholm, D. J., and Lazarus, 140-165. L. (1970). Serum gastrin in patients with peptic McFarland, R. J., Buchanan, K. D., Sloan, J. M., and ulceration. British Medical Journal, 2, 626-629. Ardill, J. (1978a). Idiopathic antral gastrin deficiency. Chey, W. Y., Hendricks, J., and Tai, H. H. (1977). Proceedings of 2nd International Symposium on Gastro- Plasma secretin in fasting and postprandial state in intestinal Hormones, Beitostolen, August-September man (Abst.). Gastroenterology, 72, 1156. 1978. Cowley, D. J., Dymock, 1. W., Boyes, B. E., Wilson, McFarland, R. J., Sloan, J. S., and Buchanan, K. D. R. Y., Stagg, B. H., Lewin, M. R., Polak, J. M., and (1978b). Variation in antral gastrin concentration and Pearse, A. G. E. (1973). Zollinger-Ellison syndrome the effect of gastritis. Gut, 19, A978. type 1: clinical and pathological correlations in a case. McGuigan, J. E., and Trudeau, W. L. (1973). Differences Guit, 14, 25-29. in rates of gastrin release in normal persons and Creutzfeldt, W., Arnold, R., Creutzfeldt, C., and Track patients with duodenal-ulcer disease. New England N. S. (1975). Pathomorphologic, biochemical, and Journal of Medicine, 288, 64-66. diagnostic aspects of gastrinomas (Zollinger-Ellison McLoughlin, J. C., Green, W. E. R., and Buchanan, syndrome). Human Pathology, 6, 47-76. K. D. (1978). Gastric emptying of ingested acid and Dockray, G. J. (1976). Immunochemical evidence of its effects on plasma gastrin and secretin in duodenal cholecystokinin-like peptides in brain. Nature (London), ulcer subjects. ScandinavianJournalofGastroenterology, 264, 568-570. 13, 313-319. Dockray, G. J., and Taylor, I. L. (1976). Heptadeca- Mason, J. C., Murphy, R. F., and Buchanan, K. D. peptide gastrin: measurement in blood by specific (1977). Characterisation of secretin-like peptides from

radioimmunoassay. Gastroenterology, 71, 971-977. tissue and plasma extracts using immunoaffinitycopyright. Edkins, J. S. (1905). The chemical mechanism of gastrin chromatography (Abst.). Gut, 18, A982-A983. secretion. Journal ofPhysiology, 34, 133-144. Mason, J. C., Murphy, R. F., Henry, R. W., and Ganguli, P. C., Polak, J. M., Pearse, A. G. E., Elder, Buchanan, K. D. (1979). Starvation-induced changes J. B., and Hegarty, M. (1974). Antral gastrin cell in secretin-like immunoreactivity of human plasma. hyperplasia in peptic ulcer disease. Lancet, 1, 583-586. Biochimica Biophysica Acta, 582, 322-331. Greenberg, G. R., and Bloom, S. R. (1978). Plasma Morley, J. S. (1968). Structure-function relationships secretin during fasting (Letter). Lancet, 1, 273. in gastrin-like peptides. Proceedings of the Royal

Gregory, R. A. (1974). The gastro-intestinal hormones: Society B, 179, 79-111. http://jcp.bmj.com/ a review of recent advances. Journal of Physiology, Mutt, V., and Jorpes, J. E. (1966). Secretin: isolation and 241,1-32. determination of structure. Fourth International Gregory, R. A., and Tracy, H. J. (1964). The constitution Symposium on the Chemistry of Natural Products, and properties of two gastrins extracted from hog Stockholm, June 26th-July 2nd. antral mucosa. Gut, 5, 103-117. O'Connor, F. A.. McLoughlin, J. C., and Buchanan, Gregory, R. A., and Tracey, H. J. (1975). Gastrointestinal K. D. (1977). Impaired immunoreactive secretin Hormones, pp. 13-24, ed J. C. Thompson. University of release in coeliac disease. British Medical Journal, 1, Texas Press, Austin and London. 811-812. on September 29, 2021 by guest. Protected Hacki, W. H., Bloom, S. R., Mitznegg, P., Domschke, Rhodes, R. A., Chey, W. Y., Tai, H. H., and Escoffery, R. W., Domschke, S., Belohlavek, D., Demling, L., and (1975a). Impaired release of secretin in celiac sprue Wunsch, E. (1977). Plasma secretin and pancreatic (Abst.). Clinical Research, 23, 578A. bicarbonate response to exogenous secretin in man. Rhodes, R. A., Chey, W. Y., Tai, H. H., and Tabechian, Gut, 18, 191-195. H. (1975b). Hypersecretinemia associated with renal Hansky, J. (1974). Antral-gastrin-cell hyperplasia in failure (Abst.). Clinical Research, 23, 255A. peptic ulcer disease (Letter). Lancet, 1, 1344-1345. Rehfeld, J. F. (1978). Radioimmunoassay of gastrin. Hansky, J., Soveny, C., and Korman, M. G. (1971). In Gut Hormones, pp. 145-148, ed S. R. Bloom. Effect of secretin on serum gastrin as measured by Churchill Livingstone, Edinburgh. immunoassay. Gastroenterology, 61, 62-68. Rehfeld, J. F., and Stadil, F. (1973). Gel filtration studies Henry, R. W., Flanagan, R. W. J., and Buchanan, K. D. on immunoreactive gastrin in serum from Zollinger- (1975). Secretin: a new role for an old hormone. Lancet, Ellison patients. Gut, 14, 369-373. 2, 202-203. Schaffalitzky de Muckadell, 0. B., and Fahrenkrug, J. Holohan, K. N., Murphy, R. F., Flanagan, R. W. J., (1978). Role of secretin in man. III. In Gut Hormones, Buchanan, K. D., and Elmore, D. T. (1973). Enzymic pp. 197-200, ed S. R. Bloom. Churchill Livingstone, iodination of the histidyl residue of secretin: a radio- Edinburgh. immunoassay of the hormone. Biochimica Biophysica Trudeau, W. L., and McGuigan, J. E. (1971). Relations Acta, 322, 178-180. between serum gastrin levels and rates of gastric Isenberg, J. I., Cano, R., and Bloom, S. R. (1977). hydrochloric acid secretion. New England Journal of Effect of graded amounts of acid instilled into the Medicine, 282, 408-412. J Clin Pathol: first published as 10.1136/jcp.s1-8.1.17 on 1 January 1978. Downloaded from

Pathophysiology ofgastrin and secretin 25 Vanderhaeghen, J. J., Signeau, J. C., and Gepts, W. Walsh, J. H., Isenberg, J. I., Ansfield, J., and Maxwell, (1975). New peptide in the vertebrae CNS reacting V. (1975). Clearance and acid-stimulating action of with antigastrin antibodies. Nature (London), 257, human big and little gastrins in duodenal ulcer 604-605. subjects. JournalofClinicalInvestigation, 57, 1125-1131. Walsh, J. H., and Grossman, M. 1. (1973). Circulating Yalow, R. S., and Berson, S. A. (1972). And now 'Big, gastrin in peptic ulcer disease. Mount Sinai Journal of Big gastrin'. Biochemical and Biophysical Research Medicine, 40, 374-381. Communications, 48, 391-395. copyright. http://jcp.bmj.com/ on September 29, 2021 by guest. Protected